Hi, I fixed CPU binding in this version of my scheduler bigpatch and have noted something that should be insightful even to the 'pros' like Linus or DaveM... The thing I noted is that processes migrate _far_ less than before, but load is spread out more evenly. Another thing is reduced interactive performance, but this is entirely due to the fact that reschedule_idle() is NOT doing the right thing. If you don't believe me about any of this, please try the patch (or only the goodness() thing I posted a few hours ago) and observe -- working processor binding will show the reschedule_idle() deficiencies to you in a very very convincing way :)) cheers, Rik -- the flu hits, the flu hits, the flu hits -- MORE +-------------------------------------------------------------------+ | Linux memory management tour guide. [EMAIL PROTECTED] | | Scouting Vries cubscout leader. http://www.phys.uu.nl/~riel/ | +-------------------------------------------------------------------+ - Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/mentre/smp-faq/ To Unsubscribe: send "unsubscribe linux-smp" to [EMAIL PROTECTED]
